LiuRuiLin
commited on
Commit
•
ed9b12a
1
Parent(s):
e190ff4
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,25 @@
|
|
| 1 |
---
|
| 2 |
license: unknown
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: unknown
|
| 3 |
---
|
| 4 |
+
|
| 5 |
+
- **Bits**: 4
|
| 6 |
+
- **Group Size**: 128
|
| 7 |
+
- **Damp Percent**: 0.01
|
| 8 |
+
- **Desc Act**: false
|
| 9 |
+
- **Static Groups**: false
|
| 10 |
+
- **Sym**: false
|
| 11 |
+
- **True Sequential**: false
|
| 12 |
+
- **LM Head**: true
|
| 13 |
+
- **Model Name or Path**: null
|
| 14 |
+
- **Model File Base Name**: model
|
| 15 |
+
- **Quant Method**: gptq
|
| 16 |
+
- **Checkpoint Format**: gptq
|
| 17 |
+
- **Meta:**
|
| 18 |
+
- Quantizer: intel/auto-round:0.1
|
| 19 |
+
- Packer: autogptq:0.8.0.dev1
|
| 20 |
+
- Iters: 400
|
| 21 |
+
- LR: 0.0025
|
| 22 |
+
- MinMax LR: 0.0025
|
| 23 |
+
- Enable MinMax Tuning: true
|
| 24 |
+
- Use Quant Input: false
|
| 25 |
+
- Scale Dtype: torch.float16
|